Economics of Philanthropy-Evidence from Health Crowdfunding

Presentation (PDF Available) · February 2019with 72 Reads 
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
DOI: 10.13140/RG.2.2.36214.04169
Developments in Entrepreneurial Finance: Crowdfunding, Blockchain, and ICOs, DOI:10.13140/RG.2.2.36214.04169
Cite this publication
Abstract
This paper is the first comprehensive empirical study on the economics of health crowdfunding (HCF) campaigns. We examine donor characteristics, in particular whether they make purely altruistic contributions, or whether they follow certain preference patterns. Our data highlights that, on average, campaign funding goals are achieved more rapidly if the patient is a baby girl, and if campaign descriptions are more comprehensive but less technical (easier to read). Furthermore, campaigns begun around holidays are funded more quickly. We document strong and economically significantly negative donor-to-donor peer effects, whereby contributions by donors with public profiles are crowded out by the previous contributions of their peers.
Economics of Philanthropy
Evidence from Health Crowdfunding
This Version: April 27th, 2018
Juliane Proelss (Concordia University)
Denis Schweizer (Concordia University)
Tingyu Zhou (Concordia University)
2
Agenda 1 2 3 4 5
Motivation
Research Questions
Data Collection
Methodology
4
3
2
1
Empirical Analysis
5
6
Conclusion
6
Appendix
3
Reward Crowdfunding
Kickstarter, Indiegogo
Equity Crowdfunding
Wefunder, Fundersclub, Localstake
Donation Crowdfunding
Watsi, GoFundMe, & Youcaring or Kiva.org
Other
Peer-to-Peer Lending (e.g. Lending Club)
Real Estate Crowdfunding (e.g. Mosaic, Reality Mogul, Crowdstreet)
Crowdfunding 12 3 4 5 6
4
Lack of access, and inability to finance medical treatment effects life expectancy of up to
thirty-six years between developing and developed countries (WHO, 2011).
Health crowdfunding is a method to raise funds for medical expenses or treatment by a
group of mostly small donors through an open call for funding on Internet-based platforms.
HCF platforms raised more than $ 1 billion in the U.S. for medical expenses and treatments
over the one-year period from September 2015 through October 2016 (Time, 2017).
Watsi.org is the only non-profit organization that raises funds in the U.S. for patients in
developing countries.
Watsi.org competes for charitable donations against about 1.5 million other charitable
organizations in the U.S. (see Giving USA and the National Center for Charitable Statistics).
Watsi has no funding competition, all campaigns are fully funded, no fraudulent behavior
occurs, and all campaigns are structured by Watsi and their partners and thus look similar.
Motivation Health Crowdfunding (HCF) 12 3 4 5 6
5
What is Watsi? A Typical Campaign 12 3 4 5 6
Patient‘s profile
picture
Campaign
description and
timeline
Funder’s icons
Link to social
media
Funding goal
and progress
Campaign story
3
2
1
5
6
4
6
What is Watsi? A Typical Funding 12 3 4 5 6
7
What is Watsi? A Typical Funder 12 3 4 5 6
8
Research Question (1/2) 1234 5 6
Public Profile Donors are less inclined to share their private benefits
with their peers (crowding out of Public Profile Donors).
H6
The crowding-out effect is stronger for “high impact” Public Profile
Donors (“Shining Knights”).
H7
Do Watsi’s donors act from a purely altruistic standpoint?
Q1
Holidays and Sundays increase the likelihood of charitable giving and
increase funding speed.
H4
9
Research Question (2/2) 1234 5 6
What motivates donors to contribute to healthcare campaigns?
Q2
Age is negatively correlated with funding speed.
H1
Treatments for female (baby) patients are funded more quickly
H2
Sympathy & emotional contagion are positively correlated with funding speed.
H3
The severity of condition is positively correlated with funding speed.
H5
10
Watsi Data Collection
Detailed costs including operational expenses that are covered by a
foundation or by donors
Link to campaign webpage URL
4,677 distinct patient campaigns
Patient Characteristics including picture, patient story, campaign dates,
funding structure
5,314 unique registered donors (Public Profile Picture (PPP) donors,
Public Profile Initials (PPI) donors, anonymous donors)
Number of funded campaigns
Funding sequence
Webpage
URL
Transparency
Document
123456
Registered
donors
11
Watsi Funded Campaigns*123456
Countries
2012 Vol. 2013 Vol. 2014 Vol. 2015 Vol. 2016 Vol. Total:
Total Vol.:
Burma
0 - 27 38,390 64 72,765 98 142,520 19 28,485 208 282,160
Burundi
0 - 0 - 0 - 1 1,260 0 - 1 1,260
Cambodia
4 1,245 128 40,200 546 127,555 687 161,646 116 28,889 1,481 359,535
Ethiopia
4 5,985 35 34,300 21 21,025 25 29,365 8 10,265 93 100,940
Ghana
0 - 1 1,500 1 100 4 6,000 0 - 6 7,600
Guatemala
22 14,000 63 50,030 170 129,550 233 155,708 27 17,311 515 366,599
Haiti
0 - 11 12,520 118 153,020 115 161,229 7 10,500 251 337,269
Kenya
0 - 104 74,655 346 214,230 378 295,515 34 25,400 862 609,800
Malawi
0 - 22 20,045 95,519 15 13,283 0 - 46 38,847
Mali
0 - 11 9,760 5 3,475 3 2,286 0 - 19 15,521
Nepal
12 11,530 47 51,220 126 54,720 91 24,353 27 10,195 303 152,018
Nigeria
0 - 3 4,200 1 1,500 7 10,403 0 - 11 16,103
Panama
0 - 1 1,400 0 - 4 4,694 0 - 5 6,094
Philippines
0 - 9 6,465 19 9,606 40 30,293 3 3,988 71 50,352
Somalia
0 - 0 - 0 - 5 3,721 0 - 5 3,721
Somaliland
0 - 1 640 31,950 88,607 0 - 12 11,197
Tanzania
0 - 79 52,775 169 142,370 208 188,292 27 27,224 483 410,661
Thailand
0 - 11 12,210 15 20,555 26 36,315 1 1,500 53 70,580
Uganda
0 - 2 3,000 26 7,825 134 31,704 65 14,580 227 57,109
Zambia
0 - 21 26,500 4 5,000 0 - 0 - 25 31,500
Total
42 32,760 576 439,810 1,643 970,765 2,082 1,307,194 334 178,337 4,677 2,928,866
*June 2012 through February 2016
12
Summary Statistics
Variable
# Obs. Mean Std. Dev. Min Max
Dependent Variables
Time
4,677 3.94 4.29 1 69
D_Time_2
4,677 0.52 0.50 0 1
D_Time_4
4,677 0.69 0.46 0 1
Patient Characteristics
Age
4,677 26.84 25.00 0 90
Baby
4,677 0.09 0.29 0 1
Child
4,677 0.39 0.49 0 1
Adult
4,677 0.36 0.48 0 1
Elder
4,677 0.16 0.36 0 1
Female
4,677 0.55 0.50 0 1
Smile
4,677 0.53 0.50 0 1
Treatment Cost
4,677 6.13 0.82 4.25 8.01
Date of Campaign Posting
First Business Day of the Month
4,677 0.15 0.35 0 1
Holiday
4,677 0.24 0.43 0 1
Weekend
4,677 0.16 0.37 0 1
Patient Story
Length
4,677 5.11 0.30 4.09 6.10
# Life
-Threatening Words 4,677 0.35 0.95 0 13
ARI
4,677 10.47 1.75 4.6 17.8
CL
4,677 11.31 1.58 5.5 19.66
Gunning Fog
4,677 8.56 1.06 5.2 14.4
FKG
4,677 8.12 1.54 2.9 15.4
FRE
4,677 66.11 8.57 29.86 94.15
123456
14
Determinants of Funding Time (Poisson Model) 12 3 456
(1) (2) (3)
Patient Characteristics
Age
0.008***
(20.51)
Baby
-0.556*** -0.440***
(-15.25) (-10.52)
Child
-0.483*** -0.478***
(-17.63) (-17.44)
Adult
-0.197*** -0.197***
(-7.68) (-7.68)
Female
-0.032** -0.028*-0.000
(-2.01) (-1.77) (-0.00)
Baby x Female
-0.287***
(-5.32)
Smile
-0.020 -0.021 -0.023
(-1.15) (-1.16) (-1.28)
Treatment Cost
0.562*** 0.547*** 0.549***
(32.41) (31.17) (31.29)
Date of Campaign Posting
First Business Day of the Month
-0.344*** -0.344*** -0.346***
(-14.55) (-14.56) (-14.62)
Holiday
-0.296*** -0.291*** -0.290***
(-15.32) (-15.07) (-15.00)
Weekend
-0.049** -0.048** -0.049**
(-2.40) (-2.34) (-2.38)
Patient’s Story Details
Length
-0.233*** -0.232*** -0.230***
(-7.04) (-7.01) (-6.94)
# Life
-Threatening Words 0.000 0.002 0.000
(0.03) (0.27) (0.03)
Constant
-1.248*** -0.639*** -0.696***
(-5.45) (-2.83) (-3.08)
Year, Country, Medical Partner FE
Yes Yes Yes
Observations
4,677 4,677 4,677
Pseudo
R20.109 0.109 0.110
15
Determinants of Funding Time (Poisson Model) 12 3 456
(1) (2) (3) (4) (5)
Patient Story (Text Indices)
ARI
0.104***
(24.92)
CL
0.129***
(27.97)
Gunning Fog
0.050***
(7.20)
FKG
0.058***
(12.07)
FRE
-0.010***
(-11.68)
Patient Characteristics
Age
0.008*** 0.008*** 0.008*** 0.008*** 0.008***
(19.97) (19.83) (20.48) (20.62) (20.67)
Female
-0.022 -0.022 -0.030*-0.027*-0.027*
(-1.40) (-1.38) (-1.91) (-1.71) (-1.72)
Smile
-0.016 -0.019 -0.019 -0.019 -0.020
(-0.92) (-1.08) (-1.07) (-1.09) (-1.16)
Treatment Cost
0.537*** 0.539*** 0.557*** 0.558*** 0.561***
(30.94) (30.96) (32.17) (32.22) (32.37)
Date of Campaign Posting
First Business Day of the Month
-0.345*** -0.350*** -0.343*** -0.352*** -0.355***
(-14.58) (-14.83) (-14.49) (-14.88) (-15.01)
Holiday
-0.283*** -0.280*** -0.295*** -0.294*** -0.294***
(-14.67) (-14.52) (-15.27) (-15.21) (-15.23)
Weekend
-0.052** -0.047** -0.052** -0.050** -0.048**
(-2.51) (-2.28) (-2.51) (-2.43) (-2.33)
Patient Story (Other)
Length
-0.214*** -0.220*** -0.228*** -0.229*** -0.233***
(-6.46) (-6.66) (-6.86) (-6.91) (-7.03)
# Life
-Threatening Words -0.002 -0.001 -0.001 -0.001 -0.001
(-0.23) (-0.08) (-0.07) (-0.16) (-0.13)
Constant
-2.357*** -2.710*** -1.687*** -1.737*** -0.592**
(-10.12) (-11.53) (-7.13) (-7.48) (-2.52)
Year, Country, Medical Partner FE
Yes Yes Yes Yes Yes
Observations
4,677 4,677 4,677 4,677 4,677
Pseudo
R20.130 0.136 0.111 0.114 0.114
17
Peer Effects on Contributions by
PP Donors
12 3 456
LPM with FE Dynamic GMM
(1) (2) (3) (4) (5) (6) (7) (8) (9)
Last Donor Public Profile
-
0.036***
(-5.64)
Min One Previous Public Profile Donor
-
0.215*** -0.527***
(-14.70) (-19.78)
Mean Impact Previous Donors
-
0.001***
-
0.002***
-
0.001*** -0.001 -0.004*** -0.001
(-10.68) (-21.69) (-19.04)
-
(-2.70) (-0.95)
Public Profile Donor Impact Higher Than Average
-
0.471*** -0.594***
(-68.75) (-9.93)
Public Profile Donor Impact Higher Than Last
-
0.324***
-
0.178***
(-64.03) (-3.11)
Campaign FE
Yes Yes Yes Yes Yes N.A. N.A. N.A. N.A.
Donation Sequence FE
Yes Yes Yes Yes Yes Yes Yes Yes Yes
Observations
51,281 51,281 51,281 51,281 51,281 37,179
37,179 37,179
R
20.275 0.279 0.276 0.440 0.345
Arellano
-Bond test for AR (1), p-value 0.000 0.000 0.006 0.000
Arellano
-Bond test for AR (2), p-value 0.603 0.736 0.235 0.363
Hansen test,
p-value 0.498 0.704 0.744 0.753
18
Peer Effects on Contributions by
PPnoM Donors
12 3 456
LPM with FE Dynamic GMM
(1) (2) (3) (4) (5) (6) (7) (8) (9)
Last Donor Public Profile
-0.082***
(0.006)
Min One Previous Public Profile Donor
-0.187*** -0.391***
(0.014) (-11.94)
Mean Impact Previous Donors
-0.000** -0.001*** -0.001*** -0.011*-0.014*** -0.016***
(0.000) (0.000) (0.000) (-1.94) (-2.98) (-2.96)
Public Profile Donor Impact Higher Than Average
-0.427*** -0.535***
(0.008) (-10.90)
Public Profile Donor Impact Higher Than Last
-0.282*** -0.335***
(0.007) (-4.33)
Campaign FE
Yes Yes Yes Yes Yes N.A. N.A. N.A. N.A.
Donation Sequence FE
Yes Yes Yes Yes Yes Yes Yes Yes Yes
Observations
51,281 51,281 51,281 51,281 51,281 37,179 37,179 37,179 37,179
R
20.216 0.215 0.211 0.348 0.264
Arellano
-Bond test for AR (1), p-value 0.000 0.000 0.001 0.000
Arellano
-Bond test for AR (2), p-value 0.288 0.313 0.13 0.383
Hansen test,
p-value 0.592 0.672 0.268 0.034
19
Peer Effects on Contributions by
PPP Donors
12 3 456
LPM with FE Dynamic GMM
(1) (2) (3) (4) (5) (6) (7) (8) (9)
Last Donor Public Profile Picture
-0.003
(-0.68)
Min One Previous Public Profile Picture Donor
-0.048*** -0.973***
(-4.52) (-73.79)
Mean Impact Previous Public Profile Picture Donors
-0.001*** -0.001*** -0.001*** 0.000 -0.000 0.000
(-9.57) (-10.11) (-9.88) (0.22) (-0.26) (0.26)
Public Profile Picture Donor Impact Higher Than Average
-0.022*** -0.068**
(-4.48) (-2.03)
Public Profile Picture Donor Impact Higher Than Last
-0.014***
-
0.183**
(-2.78) (-2.44)
Campaign FE
Yes Yes Yes Yes Yes N.A. N.A. N.A. N.A.
Donation Sequence FE
Yes Yes Yes Yes Yes Yes Yes Yes Yes
Observations
51,281 51,281 51,281 51,281 51,281 37,179
37,179 37,179
R
20.152 0.152 0.154 0.155 0.155
Arellano
-Bond test for AR (1), p-value 0.000 0.000 0.000 0.076
Arellano
-Bond test for AR (2), p-value 0.049 0.531 0.63 0.119
Hansen test,
p-value 0.437 0.334 0.293 0.748
20
Conclusion
Donors exhibit certain patient preferences, i.e. younger patients (especially baby girls)
Comprehensive but easier to read campaign descriptions achieve faster funding on
average HCF platform providers alike should pay greater attention to campaign
descriptions, because they can be actively managed
Donors are not touched by patient campaign descriptions that indicate a patient is facing
a severe and potentially life-threatening illness Create an index about the treatment
urgency to attract funding?
Donors seem to feel more generous around public holidays, i.e. Christmas or Easter
Posting of more patient campaigns to attract funding on these dates?
Contributions by donors with public profiles are crowding out subsequent contributions
by their peers (donors with public profiles) Create a virtual collector cards for public
profile donors?
12 3 4 5 6
21
Appendix Table A1: Number of observation Appendix
This table shows how the variables “Holiday” (panel A) and “# Life-Threatening Words
(panel B) are constructed and defined in Table 1.
Panel A: Holiday Schedule
Panel B: List of Life-Threatening Words
Year Christmas Easter Thanksgiving New Year's
Day
2012 2012-12-25 2012-04-08 2012-11-22 2012-01-01
2013 2013-12-25 2013-03-31 2013-11-28 2013-01-01
2014 2014-12-25 2014-04-20 2014-11-27 2014-01-01
2015 2015-12-25 2015-04-05 2015-11-26 2015-01-01
2016 2016-12-25 2016-03-27 2016-11-24 2016-01-01
Die, Death, Kill, Killer, Cancer, Cancerous, Life
-Threatening/Life threatening, Survive, Lose, Loss, Loss of body
part, Disability, Immobility
22
Appendix Table A2: Funding Period (Logit Model) Appendix
(1) (2) (3) (4) (5) (6)
Patient Characteristics
Age
-0.017*** -0.018***
(-9.76) (-9.42)
Baby
1.250*** 1.118*** 1.367*** 1.249***
(7.76) (5.94) (7.64) (6.07)
Child
1.033*** 1.027*** 1.111*** 1.106***
(8.56) (8.50) (8.13) (8.09)
Adult
0.425*** 0.424*** 0.432*** 0.432***
(3.85) (3.85) (3.44) (3.44)
Female
0.097 0.092 0.061 0.081 0.077 0.049
(1.43) (1.36) (0.86) (1.10) (1.05) (0.63)
Smile
-0.001 0.006 0.009 0.027 0.049 0.051
(-0.02) (0.08) (0.11) (0.33) (0.58) (0.60)
Treatment Cost
-1.198*** -1.159*** -1.161*** -1.406*** -1.365*** -1.367***
(-15.08) (-14.56) (-14.59) (-16.00) (-15.38) (-15.41)
Date of Campaign Posting
1
st Business Day of the Month 0.850*** 0.852*** 0.854*** 0.604*** 0.613*** 0.615***
(8.60) (8.61) (8.62) (5.63) (5.69) (5.71)
Holiday
0.655*** 0.646*** 0.645*** 0.775*** 0.763*** 0.762***
(8.50) (8.39) (8.37) (8.62) (8.49) (8.48)
Weekend
-0.056 -0.057 -0.056 0.112 0.113 0.114
(-0.63) (-0.64) (-0.63) (1.14) (1.15) (1.16)
Patient Story
Length
0.913*** 0.911*** 0.908*** 0.807*** 0.807*** 0.804***
(6.31) (6.29) (6.27) (5.16) (5.16) (5.14)
# Life
-Threatening Words 0.063*0.059 0.061*0.041 0.038 0.040
(1.75) (1.63) (1.67) (1.07) (1.00) (1.05)
Constant
3.355*** 1.981** 2.035** 6.838*** 5.362*** 5.422***
(3.38) (2.04) (2.09) (6.14) (4.90) (4.95)
Year FE
Yes Yes Yes Yes Yes Yes
Country FE
Yes Yes Yes Yes Yes Yes
Medical Partner FE
Yes Yes Yes Yes Yes Yes
Observations
4,667 4,667 4,667 4,667 4,667 4,667
Pseudo
R20.119 0.119 0.119 0.140 0.141 0.141
23
Appendix Table A3: Funding Period (Logit Model) Appendix
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Patient Story Details (Text Indices)
ARI -0.199*** -0.248***
(-10.49) (-11.86)
CL -0.174*** -0.322***
(-8.40) (-13.73)
Gunning Fog -0.218*** -0.093***
(-7.09) (-2.82)
FKG -0.027 -0.132***
(-1.27) (-5.76)
FRE -0.008** 0.024***
(-2.25) (5.89)
Patient Characteristics
Age -0.017*** -0.017*** -0.017*** -0.017*** -0.017*** -0.018*** -0.018*** -0.018*** -0.018*** -0.018***
(-9.65) (-9.63) (-9.75) (-9.76) (-9.74) (-9.29) (-9.27) (-9.40) (-9.45) (-9.48)
Female 0.080 0.086 0.089 0.095 0.100 0.060 0.058 0.078 0.070 0.070
(1.17) (1.26) (1.31) (1.40) (1.48) (0.80) (0.77) (1.05) (0.95) (0.94)
Smile -0.001 0.001 -0.005 -0.001 -0.002 0.018 0.028 0.024 0.025 0.028
(-0.01) (0.01) (-0.06) (-0.02) (-0.03) (0.22) (0.33) (0.29) (0.29) (0.33)
Treatment Cost -1.179*** -1.185*** -1.190*** -1.196*** -1.201*** -1.392*** -1.409*** -1.400*** -1.405*** -1.411***
(-14.66) (-14.81) (-14.89) (-15.05) (-15.10) (-15.57) (-15.65) (-15.92) (-15.94) (-16.01)
Date of Campaign Posting
First Business Day of the Month 0.876*** 0.868*** 0.861*** 0.855*** 0.840*** 0.632*** 0.654*** 0.603*** 0.630*** 0.640***
(8.74) (8.69) (8.68) (8.64) (8.48) (5.77) (5.90) (5.62) (5.85) (5.93)
Holiday 0.646*** 0.644*** 0.657*** 0.653*** 0.658*** 0.771*** 0.773*** 0.774*** 0.773*** 0.774***
(8.27) (8.29) (8.48) (8.48) (8.53) (8.45) (8.41) (8.61) (8.58) (8.58)
Weekend -0.038 -0.054 -0.037 -0.055 -0.056 0.130 0.116 0.118 0.117 0.111
(-0.42) (-0.60) (-0.41) (-0.61) (-0.63) (1.31) (1.16) (1.21) (1.19) (1.13)
Patient Story Details (Other)
Length 0.907*** 0.920*** 0.902*** 0.912*** 0.913*** 0.802*** 0.830*** 0.800*** 0.808*** 0.816***
(6.17) (6.29) (6.20) (6.30) (6.31) (5.03) (5.17) (5.11) (5.14) (5.19)
# Life-Threatening Words 0.067*0.063*0.068*0.064*0.063*0.045 0.043 0.042 0.043 0.043
(1.85) (1.74) (1.88) (1.77) (1.73) (1.18) (1.12) (1.11) (1.15) (1.13)
Constant 5.439*** 5.231*** 5.293*** 3.567*** 3.937*** 9.531*** 10.550*** 7.651*** 7.915*** 5.219***
(5.31) (5.10) (5.11) (3.55) (3.84) (8.23) (8.94) (6.64) (6.99) (4.55)
Year FE Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Country FE Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Medical Partner FE Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Observations 4,667 4,667 4,667 4,667 4,667 4,667 4,667 4,667 4,667 4,667
Pseudo R20.137 0.130 0.127 0.119 0.120 0.166 0.175 0.142 0.146 0.146
  • Article
    Full-text available
    Entrepreneurial finance markets are in a dynamic state. New market niches and players have developed and continue to emerge. The rules of the game and the methods for receiving financial backing have changed in many ways. This editorial and the special issue of Small Business Economics focus on crowdfunding (CF) and initial coin offerings (ICOs), which are two distinct but important entrepreneurial finance market segments of the future. Although the two market segments initially appear to be similar, we identify differences between them. Our comparison focuses on the stakeholders, microstructures, regulatory environments, and development of the markets. We conclude with suggestions for future ICO and CF research.
  • Life-Threatening/Life threatening, Survive, Lose, Loss, Loss of body part
    • Die
    • Death
    • Kill
    • Killer
    • Cancerous Cancer
    Die, Death, Kill, Killer, Cancer, Cancerous, Life-Threatening/Life threatening, Survive, Lose, Loss, Loss of body part, Disability, Immobility